linux-nfs.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "bfields@fieldses.org" <bfields@fieldses.org>
To: Trond Myklebust <trondmy@hammerspace.com>
Cc: "linux-nfs@vger.kernel.org" <linux-nfs@vger.kernel.org>
Subject: Re: [PATCH 13/15] knfsd: Simplify NFS duplicate replay cache
Date: Wed, 3 Oct 2018 14:11:50 -0400	[thread overview]
Message-ID: <20181003181150.GH17517@fieldses.org> (raw)
In-Reply-To: <15f0759d1fd9b6442d8634b1cb46fb59cbdf8c95.camel@hammerspace.com>

On Wed, Oct 03, 2018 at 06:01:36PM +0000, Trond Myklebust wrote:
> On Wed, 2018-10-03 at 13:14 -0400, J . Bruce Fields wrote:
> > On Mon, Oct 01, 2018 at 10:41:55AM -0400, Trond Myklebust wrote:
> > > Simplify the duplicate replay cache by initialising the
> > > preallocated
> > > cache entry, so that we can use it as a key for the cache lookup.
> > > 
> > > Note that the 99.999% case we want to optimise for is still the one
> > > where the lookup fails, and we have to add this entry to the cache,
> > > so preinitialising should not cause a performance penalty.
> > > 
> > > Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
> > > ---
> > >  fs/nfsd/cache.h    |  6 +--
> > >  fs/nfsd/nfscache.c | 94 ++++++++++++++++++++++------------------
> > > ------
> > >  2 files changed, 47 insertions(+), 53 deletions(-)
> > > 
> > > diff --git a/fs/nfsd/cache.h b/fs/nfsd/cache.h
> > > index b7559c6f2b97..bd77d5f6fe53 100644
> > > --- a/fs/nfsd/cache.h
> > > +++ b/fs/nfsd/cache.h
> > > @@ -24,13 +24,13 @@ struct svc_cacherep {
> > >  	unsigned char		c_state,	/* unused, inprog,
> > > done */
> > >  				c_type,		/* status, buffer
> > > */
> > >  				c_secure : 1;	/* req came from
> > > port < 1024 */
> > > -	struct sockaddr_in6	c_addr;
> > >  	__be32			c_xid;
> > > -	u32			c_prot;
> > > +	__wsum			c_csum;
> > >  	u32			c_proc;
> > > +	u32			c_prot;
> > >  	u32			c_vers;
> > >  	unsigned int		c_len;
> > > -	__wsum			c_csum;
> > > +	struct sockaddr_in6	c_addr;
> > >  	unsigned long		c_timestamp;
> > >  	union {
> > >  		struct kvec	u_vec;
> > 
> > Unless I've missed something subtle--I'll move this chunk into the
> > next
> > patch.--b.
> 
> Nothing too subtle. The only reason for keeping it in this patch is
> because even with the current code, most of the comparisons hit the
> c_xid and possibly sometimes the c_csum, so those are the main fields
> that you want to try to keep in the same cache line.

That could use a comment.

--b.

  reply	other threads:[~2018-10-04  1:01 UTC|newest]

Thread overview: 26+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-10-01 14:41 [PATCH 00/15] Performance improvements for knfsd Trond Myklebust
2018-10-01 14:41 ` [PATCH 01/15] SUNRPC: Remove the server 'authtab_lock' and just use RCU Trond Myklebust
2018-10-01 14:41   ` [PATCH 02/15] SUNRPC: Add lockless lookup of the server's auth domain Trond Myklebust
2018-10-01 14:41     ` [PATCH 03/15] SUNRPC: Allow cache lookups to use RCU protection rather than the r/w spinlock Trond Myklebust
2018-10-01 14:41       ` [PATCH 04/15] SUNRPC: Make server side AUTH_UNIX use lockless lookups Trond Myklebust
2018-10-01 14:41         ` [PATCH 05/15] knfsd: Allow lockless lookups of the exports Trond Myklebust
2018-10-01 14:41           ` [PATCH 06/15] SUNRPC: Lockless server RPCSEC_GSS context lookup Trond Myklebust
2018-10-01 14:41             ` [PATCH 07/15] knfsd: Lockless lookup of NFSv4 identities Trond Myklebust
2018-10-01 14:41               ` [PATCH 08/15] NFS: Lockless DNS lookups Trond Myklebust
2018-10-01 14:41                 ` [PATCH 09/15] SUNRPC: Remove non-RCU protected lookup Trond Myklebust
2018-10-01 14:41                   ` [PATCH 10/15] SUNRPC: Replace the cache_detail->hash_lock with a regular spinlock Trond Myklebust
2018-10-01 14:41                     ` [PATCH 11/15] SUNRPC: Simplify TCP receive code Trond Myklebust
2018-10-01 14:41                       ` [PATCH 12/15] knfsd: Remove dead code from nfsd_cache_lookup Trond Myklebust
2018-10-01 14:41                         ` [PATCH 13/15] knfsd: Simplify NFS duplicate replay cache Trond Myklebust
2018-10-01 14:41                           ` [PATCH 14/15] knfsd: Further simplify the cache lookup Trond Myklebust
2018-10-01 14:41                             ` [PATCH 15/15] knfsd: Improve lookup performance in the duplicate reply cache using an rbtree Trond Myklebust
2018-10-04  0:44                               ` J . Bruce Fields
2018-10-03 17:14                           ` [PATCH 13/15] knfsd: Simplify NFS duplicate replay cache J . Bruce Fields
2018-10-03 18:01                             ` Trond Myklebust
2018-10-03 18:11                               ` bfields [this message]
2018-10-03 23:51                                 ` bfields
2018-10-03 16:10                 ` [PATCH 08/15] NFS: Lockless DNS lookups J . Bruce Fields
2018-10-03 17:55                   ` Trond Myklebust
2018-10-03 16:08       ` [PATCH 03/15] SUNRPC: Allow cache lookups to use RCU protection rather than the r/w spinlock J . Bruce Fields
2018-10-02 19:39   ` [PATCH 01/15] SUNRPC: Remove the server 'authtab_lock' and just use RCU J . Bruce Fields
2018-10-01 15:29 ` [PATCH 00/15] Performance improvements for knfsd Trond Myklebust

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20181003181150.GH17517@fieldses.org \
    --to=bfields@fieldses.org \
    --cc=linux-nfs@vger.kernel.org \
    --cc=trondmy@hammerspace.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).